cudnn via pip

README.md

@@ -24,7 +24,7 @@ conda activate cuda11.2.2_py3.10
 
 ### Install conda packages
 ```sh
-conda install -c conda-forge cudatoolkit=x.y.z cudnn=a.b.c
+conda install -c conda-forge cudatoolkit=x.y.z
 conda install -c nvidia cuda-nvcc
 ```
 The correct package versions can be found in the [tensorflow installation guide](https://www.tensorflow.org/install/pip).
@@ -59,7 +59,7 @@ source venv/bin/activate
 Then install tensorflow (compatible package in the [tensorflow installation guide](https://www.tensorflow.org/install/pip)) and [tensorrt](https://docs.nvidia.com/deeplearning/tensorrt/)
 ```sh
 pip install nvidia-pyindex
-pip install tensorflow==x.y.z nvidia-tensorrt
+pip install tensorflow==x.y.z nvidia-tensorrt nvidia-cudnn-cuxx==a.b.c.def
 ```
 
 Unfortunately, if tensorrt 7.\*.\*.\* is needed (not available on pypi), you must manually link the installed `libnvinver` and `libinver_plugin` to version 7, i.e.